Research On Task Scheduling Technology For Computing Offloading In Mobile Terminal Distributed Networks

Mobile terminal distributed network is generally composed of reciprocal terminal devices,and each node undertakes task processing and task forwarding.How to make full use of the limited energy of device,ensure the smooth operation of network,and improve the efficiency of task processing are the key technical problems in this field.Existing researches generally adopt centralized algorithm for task scheduling,and divide the network into multiple regional subcenters,which makes it difficult to make full use of the overall resources of equipment.Computing offloading technology focuses on the transfer of computing tasks from high-loading nodes to low-loading nodes,so as to improve the utilization level of resources in the whole network.Efficient task scheduling method is very important for extending the lifetime of terminal distributed network and reducing the time delay of task processing by using computing offloading technology.Therefore,this thesis focuses on the computing offloading task scheduling technology in the mobile terminal distributed network and carries out the following work:(1)A compensational computation offloading algorithm to maximize the lifetime of terminal distributed network under constrained energy is proposed.By modeling the distributed network,the optimal offloading scheme is determined by using the minimum cost maximum flow algorithm in graph theory,and then the approximate solution is obtained by iterative method.The ultimate goal of this algorithm is to adjust the computing energy consumption and transmission energy consumption of each node by task scheduling,and finally achieve the lifetime balance of the whole network,ensure the quality of service.Simulation results show that the proposed algorithm can increase the original network life time to 1 70?240%.(2)A throughput constrained computation offloading algorithm based on terminal distributed network under data constrained is proposed.This algorithm is based on the computing offloading in the terminal distributed network,and uses the dynamic programming and knapsack theory to determine the best offloading scheme,so as to transfer the high-delay tasks of some terminals to their adjacent terminals for execution,reduce the overall task processing delay,improve the response speed,optimize the user experience,and effectively alleviate the congestion of some nodes.Simulation results show that the proposed algorithm can reduce the original network response time delay by 20-40%and improve the system performance.The computing offloading scheduling technology based on mobile terminal distributed network proposed in this thesis has certain research and application value in improving the overall lifetime of the network,reducing the delay of task processing and improving the response speed,which can provide certain research and reference significance for related distributed network computing offloading research.